Sunday, May 30, 2010

Processor artwork

Because i had a lot of processors and microcontrollers laying around, i made some "artwork" out of the stuff.
I bought a piece of ESD foam and sticked and glued all the stuff onto it.

Partlist:

Microprocessors:
  • Harris 80C286, 16 MHz
  • Cyrix 80387 math. co-processor clone, 33 MHz
  • Intel 80386DX, 33 MHz
  • Intel Pentium 1, 166 MHz
  • Cyrix Media GXm, 266 MHz
  • AMD K6-III, 400 MHz
  • AMD K6 2, 500 MHz
 Microcontrollers:
  • NEC 8085AHC
  • Siemens 80C537, 16 MHz
  • Atmel Atmega 8, 16 MHz
The two silver pieces in the middle of the foam are hard disk magnets, holding the foam on the magnet board.
An AMD 600 MHz Athlon slot a was to heavy to be glued on the foam. ;)

HDD storage tower made out of Meccano like parts

The idea was not new, but a few days ago i had the time and passion to make the dream come true.
There are five PATA hard disks mounted in the construction. The construction is quite stable and looks pretty.
I am sure: Gustave Eiffel would like it.
The reason for this construction was the bad hard disk alignment in this chieftech big tower. For SATA it is ok, but for PATA disks the slots are not very well suited.

Unfortunately the system disk of the server died while i was trying to figure out the correct PATA cable and connector assignment. (I had to reconnect the disk exactly as they were connected before the modification)
The system disk was not mounted in the tower - so it was some kind of natural selection not my own stupidy. ;)
Watch the video for the cool sound the disk makes...








p.s.: i found a guy who made a fan holder out of meccano: http://img.techpowerup.org/070507/2.jpg.

Temperature sensor disassembled



Some days ago i disassembled an RF temperature sensor used for a weather station.
It sends on 433 MHz.
Every 90 seconds the sensor is sending the actual temperature to the station(s).













Powered by two AAA batteries.
The chassis is sealed.
















The electronic circuit is rather small.
There are two PCBs. On for the RF part i guess with a big coil.
The other PCB contains the controller. It is covered by a black glue.
There is a contact on the PCB for the switch in the chassis for choosing the channel, marked with "TX" (by pressing it, the sensor is sending the temperature immediately).
The other contact is not brought out to the chassis. It is for switching between °C and °F.










This it the other side of the PCBs. You can see the red LED, that will blink every time the temperature was measured and send.
The contacts above the LED are looking interesting. Perhaps they were used for programing or testing the circuit?



And of course the sensor worked again after i reassemled it. ;)

Saturday, May 15, 2010

Finally!

I was able to get my old Hewlett Packard photosmart 1115 to work - with Windows Vista.
It took me 8 hours to plan and setup an external print server with linux and cups.
Finding the right operating system was quite a challenge.

But now the s*** works!
In the end i installed Debian Linux 5.
And i'm still a bit confused why Vista is so stubborn when it comes to printer support. (And HP in supporting old printers with drivers for new operating systems)
With my old XP i have no problems - no matter how i connect the printer to the computer - the printer just works.
The opposite with Vista - no matter how i tried getting it to work - it was resisting my efforts... till today.
But now: the resistance is futile! XD

Saturday, April 24, 2010

Thermometer version 5

I added backlight functionality to my thermometer.
A menu system was added too (as well as an annoying bug...).

I noticed that the button response is getting worse with time and the number of button clicks. Perhaps they are just to cheap? It is not a software fault.
The possibility to calibrate the clock for higher accuracy is still missing.

Gameport to PS/2 converter




Finally! My gameport project is finished!
With this little circuit it is possible to emulate a ps/2 keyboard with your gamepad.
I designed the circuit for a game that could not be played with that gamepad. But it worked well with a keyboard - so this evil mad idea was born. ;)
I used my old repaired gravis gamepad (repair report) for that project. Other gamepads could be wired internal in a different way.
The circuit is designed for digital gamepads only - no analog sticks or joysticks are supported (yet)!
But such a functionality could be implemented too. Because i used an old ATMEGA8, there are a lot of free pins for further experiments and hacks.
The buttons are "hard wired" in the code to keys. I used the ALT, SPACE, ENTER and CNTRL keys for the buttons. A function to define new key bindings without hacking them into the code and reporogram the controller could be implemented - but i was to lazy to write it. ^.^

Another impressive project that shows how easy and yet powerfull a µC circuit can be.
THX to the KISS guidelines.

The circuit in the red box does all the magic.












Feel free to use or improve this circuit.


The code can be found here:
http://sites.google.com/site/species0x2118/hive
http://sites.google.com/site/species0x2118/hive/gamepad.c?attredirects=0&d=1 (direct link)
I changed the fuses to run the controller with 8 MHz to have enough speed reserves. But a lower frequency could work as well.

As a goody i made a video of the gameplay with the gamepad connected to the converter:


Wednesday, April 21, 2010

Gamepad fixed


Last week i repaired my old Gravis gamepad. It had a broken cable. To fix it, i bought a joystick extension cable as replacement. Because the new cable was a bit thicker, i had to modify the casing of the gamepad. The modifications are not visible from the outside.